Populism, political risk and the economy: What we can learn from the Italian experience

Pierluigi Balduzzi, Emanuele Brancati, Marco Brianti, Fabio Schiantarelli 20 February 2020

a

A

In the years that followed the global and sovereign debt crises, Italy has experienced a rise of populist forces (Lega and Movimento 5 Stelle) that culminated in June 2018 in the formation of a populist government, which lasted approximately one year. These political developments affected the concerns and risk associated with budgetary policies, debt sustainability, and the very prospects of continued membership of the euro.

What have been the effects of political risk shocks during the period 2013 to 2019 on Italy’s domestic financial markets and real economy? Are there any spillover effects to financial markets of other euro area countries?  These questions are of primary importance in light of the rise of populist forces in other countries, albeit with their own specificities, that are profoundly altering the political landscape.1

Not surprisingly, all these international developments have generated an increase in economic policy uncertainty, and a series of papers provide empirical evidence on its adverse effects on the financial markets and the real economy. For instance, Baker et al. (2016) analyse the effects of economic policy uncertainty on the US economy. Kelly et al. (2016) focus instead on the effect of political uncertainty around election and international summits on stock option contracts. Caldara et al. (2019) and Baker et al. (2019a) discuss the consequences of trade uncertainty for US firms and for the US stock market, respectively. Finally, Bloom et al. (2019) analyse the implications of Brexit for UK firms.2

One element that is interesting and distinctive about the Italian case is that the rise in populism occurred in the context of a high level of government debt (around 130% of GDP) and decades of slow growth. These factors contributed to concerns about debt sustainability. Thus, the Italian experience is very relevant for countries in similar circumstances. However, Italy’s economic importance (it is the third largest euro area country and the eight largest in the world in terms of GDP) makes the Italian case of general significance, as domestic political risk shocks have the potential to generate direct and indirect spillover effects to other euro area countries. 

A quick look at the evolution of the sovereign credit default swap (CDS) spreads for the 2014 clause dollar-denominated contracts from September 2014 onward suggests that many of the significant fluctuations occur as a result of important domestic political developments.3 These can be seen in Figure 1, where we report Italy’s sovereign CDS spreads together with those of other euro area countries. In particular, the CDS spread displays a pronounced spike between the end of the second quarter and the beginning of the third quarter of 2018, reaching values above 250 basis points – an upward jump of around 150 basis points. In addition, we observe smaller increases at the same time in the sovereign CDS spreads of other euro area countries. This upward movement coincides with a series of events which saw the two major populist parties in Italy – Lega and Movimento 5 Stelle – creating first a coalition and eventually a government. Note also that the sovereign CDS spread for Italy has been much higher, particularly since 2018, compared not only to the spreads for Germany and France, but also to Ireland, Spain and Portugal. 

Figure 1 Sovereign CDS spread, 2014 clause contracts, for selected euro area countries

Note: The figure reports the spread on the dollar-denominated 2014 clause 5-year sovereign CDS contracts for France, Germany, Ireland, Italy, Portugal and Spain.

In a new paper (Balduzzi et al. 2020), we build on these observations and assume that the daily change in the Italian sovereign CDS spread on the dates of political events (e.g. elections and the formation  of a government) and policy announcements (e.g. the submission of the draft budget to the European Commission and the subsequent correspondence) is informative about the unobserved shocks to concerns associated with budgetary policies, government debt sustainability, and membership of the euro in Italy.4 This is a plausible hypothesis, as the sovereign CDS spread reflects the probability of the government defaulting on its debt and the expected losses for bond holders in that case. We apply the methodology discussed in Stock and Watson (2018) and use the daily change in the CDS spread for Italian government bonds on political and policy announcement dates as an instrument for political risk shocks in the context of local projections (Jorda 2005). The choice of a high-frequency change in the CDS, together with the use of controls for international risk, is made in order to plausibly satisfy the exogeneity assumptions required by the local projection–instrumental variables (LP-IV) procedure.5

The change in the CDS spreads on our selected dates (see Figure 2) highlights the importance of the rise to power of populist forces and of their policy announcements in generating an increase in political risk. At the same time, it also emphasises the relevance of institutional constraints such as the European Commission and the Italian presidency placing limits on budgetary policies perceived as risky and on a repositioning of Italy with respect to the European fiscal rules and membership of the common currency. These constraints triggered risk-decreasing shocks that cushioned the increase in all the spreads and the negative effects on the stock market stemming from the rise and ascendance to power of the populist forces.6 

Figure 2 ∆Sovereign CDS spread, 2014 clause contracts, around political events

Note: Changes in the CDS spread of the sovereign 2014-clause contract denominated in dollars around selected dates. Changes are defined as the closing price of the event day minus the closing price of the previous day. EC stands for European Commission.

We use the LP-IV approach, with the instrument obtained using the high-frequency identification strategy described above, to analyse the effect of political risk shocks on financial markets using data at a daily and monthly frequency. Our evidence from the response of each endogenous variable, at different horizons, to a single shock indicates that increases in political risk associated with the rise to power of populist forces have a powerful effect on both domestic and international financial variables. Italian sovereign and bank CDS spreads, as well as the Italian BTP-German Bund yield spreads, increase, making government and bank borrowing more expensive. For a sense of the size of the quantitative impact, note that a unit change in our instrument leads to a sustained change of the BTP-Bund spread that is twice as large.7 Political risk shocks also negatively affect Italian stock market returns, although the effect is not as pronounced as for the spreads. 

We also provide evidence that political risk shocks affect the difference between the spread for the 2014 and the 2003 clause Italian CDS contracts denominated in the same currency (defined as the ISDA basis), which captures the probability of redenomination and depreciation of the new currency in that case. Our shocks also affect the quanto spread, defined here as the difference between the spreads on the dollar-denominated and euro-denominated Italian sovereign CDS contracts, which reflects the probability of sovereign default and the associated expected depreciation of the euro relative to the dollar. Note that the 2014 and 2003 contracts begin to diverge at the beginning of 2017, but the gap opens up dramatically in June 2018, with the former displaying much larger increases, reaching 286 basis points in mid-November 2018, while the latter increased only to 177 basis points.8

Importantly, the sovereign CDS spreads for Spain and Portugal display significant and sizable responses to Italian political risk shocks. There is also an effect on France and Germany, but this is smaller and less significant. Similarly, in terms of the spread of government bond yields relative to the German Bund, the quantitative effects are larger for Spain and Portugal, followed by Ireland and France. 

Finally, we show that these shocks have had adverse effects on the real economy and present some evidence supporting this conclusion based on the monthly Purchasing Managers Index (PMI) and other leading indicators of real activity from the beginning of 2013 until July 2019. The negative effects, although significant, were not as large as one might have feared. This is, most likely, because the political shocks we have analysed occurred in the context of a large degree of monetary accommodation and the provision of ample liquidity by the ECB. This has contributed to preventing Italian spreads from reaching the levels observed during the sovereign debt crisis in 2011-2012. In addition, the strengthening of banks' balance sheets following a period of positive growth from 2015 to the middle of 2017, the recapitalisation exercises prompted by the European Banking Authority (EBA) stress tests, and the reduction in the share of non-performing loans has allowed banks to deal with the increase in the spread in government and bank bonds without significantly increasing lending rates. As a result, the bank cost of funding channel was not important during this period, while it was key during the global and sovereign debt crises, as shown in Balduzzi et al. (2018a). In sum, all the factors we have mentioned have lessened, but not eliminated, the negative impact of the rise of populism on the real economy in Italy. 

Although Lega is now out of the government that has been formed by the more conventional centre-left parties in coalition with Movimento 5 Stelle, Italy continues to be characterised by political instability. There are legitimate questions about the ability of the government to survive as well as concerns about the outcome of an election, if it were to be called, in light of the continued strength in the polls of the Lega party. In addition, there are doubts about the possibility of a political equilibrium in Italy in support of reforms that could set the country on a more favourable growth path, which would help with debt sustainability. All this is reflected in the level of the sovereign CDS spread that, although smaller than in the second half of 2018, it is still around 140 basis points – higher than that of all other euro area countries (except Greece). Italy has been an ideal laboratory to explore and learn about the economic consequences of political risk shocks, and this is likely to continue to be the case in the future.

References

Alfaro, I, N Bloom, and X Lin (2018), “The finance uncertainty multiplier”, NBER working paper 24571. 

Azzimonti, M (2018), “Partisan conflict and private investment”, Journal of Monetary Economics 93: 114–131. 

Balduzzi, P, E Brancati, and F Schiantarelli (2018a), “Financial markets, banks’ cost of funding, and firms’ decisions: lessons from two crises”, Journal of Financial Intermediation 36: 1-15.

Balduzzi, P, E Brancati, and F Schiantarelli (2018b), “Crowding out risk: Sovereign debt, banks, and firms’ investment in Italy”, VoxEU.org, 9 November. 

Balduzzi, P, E Brancati, M Brianti, and F Schiantarelli (2020), “Populism, political risk and the economy: Lessons from Italy”, IZA Discussion Paper 12929,and Boston College Working Paper 989. 

Baker, S R, N Bloom, and S J Davis (2016), “Measuring economic policy uncertainty”, Quarterly Journal of Economics 131(4): 1593–1636.

Baker, S, N Bloom, and S J Davis (2019a), “The extraordinary rise in trade policy uncertainty”, VoxEU.org, 17 September.

Baker, S, N Bloom, S J Davis and K Kost (2019b), “Policy News and Stock Market Volatility”, NBER working paper 25720.

Basu, S and B Bundick (2017), “Uncertainty shocks in a model of effective demand”, Econometrica 85(3): 937–958. 

Bloom, N (2009), “The impact of uncertainty shocks”, Econometrica 77(3): 623–685. 

Bloom, N, M Floetotto, N Jaimovich, I Saporta-Eksten, and S J Terry (2018), “Really uncertain business cycles”, Econometrica 86(3): 1031–1065. 

Bloom, N (2014), “Fluctuations in uncertainty”, Journal of Economic Perspectives 28(2): 153–76. 

Bloom N, P Bunn, S Chen, P Mizen, P Smietanka, G Thwaites (2019), “The impact of Brexit on UK firms”, VoxEU.org, 4 September.

Caldara, D, M Iacoviello, P Molligo, A Prestipino, and A Raffo (2019), “The economic effects of trade policy uncertainty”, Journal of Monetary Economics, forthcoming. 

Cherubini, U (2019), “The econometrics of redenomination risk”, mimeo, University of Bologna.

De Santis, R A (2019), “Redenomination risk”, Journal of Money, Credit and Banking 51: 2173-2206.

Fernàndez-Villaverde, J, P Guerròn-Quintana, K Kuester, and J Rubio-Ramìrez (2015), “Fiscal volatility shocks and economic activity”, American Economic Review 105(11): 3352–84. 

Fieldhouse, A J, K Mertens, and M O Ravn (2018), “The macroeconomic effects of government asset purchases: Evidence from postwar us housing credit policy”, Quarterly Journal of Economics 133(3): 1503–1560. 

Gertler, M and P Karadi (2015), “Monetary policy surprises, credit costs, and economic activity”, American Economic Journal: Macroeconomics 7(1): 44–76. 

Gros, D (2018), “Italian risk spread: Fiscal versus redenomination risk”, VoxEU.org, 29 August.

Jordà, O (2005), “Estimation and inference of impulse responses by local projections”, American Economic Review 95(1): 161–182. 

Kelly, B, L Pàstor, and P Veronesi (2016), “The price of political uncertainty: Theory and evidence from the option market”, Journal of Finance 71(5): 2417–2480. 

Kremens, L (2019), “Currency redenomination risk”, mimeo, University of Washington.

Leduc, S and Z Liu (2016), “Uncertainty shocks are aggregate demand shocks”, Journal of Monetary Economics 82: 20–35. 

Mertens, K and M O Ravn (2013), “The dynamic effects of personal and corporate income tax changes in the United States”, American Economic Review 103(4): 1212–47. 

Minenna, M (2017), “CDS markets signal rising fear of euro breakup”, Financial Times, 6 March. 

Pàstor, L and P Veronesi (2012), “Uncertainty about government policy and stock prices”, Journal of Finance 67(4): 1219–1264.

Pàstor, L and P Veronesi (2013), “Political uncertainty and risk premia”, Journal of Financial Economics 110(3): 520–545.

Ramey, V A and M D Shapiro (1998), “Costly capital reallocation and the effects of government spending”, in Carnegie-Rochester Conference Series on Public Policy, Volume 48, Elsevier, pp. 145–194. 

Stock, J H and M W Watson (2012), “Disentangling the channels of the 2007-2009 recession”, Brookings Papers on Economic Activity, 81–135. 

Stock, J H and M W Watson (2018), “Identification and estimation of dynamic causal effects in macroeconomics using external instruments”, Economic Journal 128(610): 917–948. 

Visco, I. (2018), “Remarks by the Governor of the Bank of Italy”, Giornata Mondiale del Risparmio, 31 October.

Endnotes

1 For instance, the US, Poland and Hungary are facing changes in their institutional fabric. In addition, the UK and the US, together with many other countries, have been radically changing their international position in terms of trade and immigration policies.

2 The papers mentioned above, with the exception of Kelly et al. (2016), use measures of uncertainty derived from newspaper coverage. Other papers on the effect of political or policy uncertainty are Azzimonti (2018), Fernàndez et al. (2015), Pastor and Veronesi (2012), Pastor and Veronesi (2013) and Baker et al. (2019b).  On the economic effect of uncertainty in general, see Bloom (2009), Leduc and Liu (2016), Basu and Bundick (2017), Bloom et al. (2018), and Alfaro et al. (2018) among others. See Bloom (2014) for a survey.

3 The 2014 clause-CDS contract considers the redenomination of the sovereign debt by a G7 country or an OECD country with an investment-grade government debt as a default event. We report the figures for the dollar denominated contract as it has the more liquid market.

4 More precisely, we focus on those dates on which general elections for the Italian and European parliament took place, as well as the dates when the President of the Italian Republic chooses a political leader to attempt forming a government (Incarico). They also include the dates in which the budget law is introduced and later revised to be sent to the European Commission for approval, as well as replies of the Commission. Finally, we consider changes around the time of a few significant announcements, such as the formation of a novel coalition between the populist parties (Movimento 5 Stelle and Lega) after the last general elections (Il Contratto), and the recent withdrawal of one of the two parties (Lega) from the government.

5 The use of external instruments in local projections is also found in Fieldhouse et al. (2018). See also Stock and Watson (2012), Mertens and Ravn (2013), Gertler and Karadi (2015), and Ramey and Shapiro (1998) for estimation of structural VARs using external instruments. The econometric methodology is described in details in Balduzzi et al. (2020). In essence, the variable being instrumented (indicator variable) is, in our case,  the change in the sovereign CDS itself at all dates, under the normalizing assumption that a change of one unit in political risk is associated with a unit change in the CDS spread. The instruments must satisfy two key exogeneity assumptions to be valid. The first one (contemporaneous exogeneity) requires that changes in sovereign CDS spread on the selected dates are orthogonal to the other structural shocks in the system occurring at the same time. This must hold conditionally on a set of controls that include in our case past values of Italian financial market variables, as well as the contemporaneous and past values of the log-change in the VIX and of the first principal component of the change in the CDS spread for euro countries (with the exception of Greece and Italy) plus the UK. The choice of a narrow window of a day and of a limited number of dates is meant to maximize the probability that this is the case. The second assumption (lead-lag exogeneity) requires our instrument to be also uncorrelated with leads and lags of all the shocks in the system. We address this issue by including as a control lagged values of the dependent variable and of the instrument itself. We also we conduct a standard placebo test where we define our instrument as the change in the CDS spread on a randomly chosen set of dates. The results are supportive of our approach. Finally we implement a large set of robustness checks, including some in which we reduce the number of dates, eliminating those around which other international shocks may have occurred. 

6  Specifically, there are drastic increases in the spread in correspondence of the announcement of the contract between the Movimento 5 Stelle and Lega parties to govern together in May, which outlined the intention of pursuing a very expansionary fiscal policy based on an increase in welfare payments (‘Reddito di Cittadinanza’) and a lowering of the retirement age (‘Quota 100’). Those concerns were enhanced by the intention of the incoming populist government to propose Paolo Savona, an economist that has expressed opposition to Italian membership of the euro, to the position of Minister of Economy and Finance (the main economic post in the Italian government). The increase in the spread that followed, was in part reversed by the opposition of the President of the Republic, Sergio Mattarella, who imposed the choice of a more moderate Minister of Economy, Giovanni Tria. Another upward movement in the spread followed the drafting of the budget law at the end of 2018 that contemplated a budget deficit of 2.4% of GDP, and its subsequent rejection by the European Commission. It also eventually triggered a procedure for excessive debt. The achievement of a compromise with a lower deficit and the end of the excessive debt procedure brought some respite. However, the cumulated effect over the period captures the market's increasing overall concerns about budgetary sustainability and Italy's position concerning fiscal rules and the Euro. 

7  A simple back-of-the-envelope way to think about the quantitative effects is to observe that the 5-years BTP-Bund spread fluctuated around an average value of 216 basis points during the populist government, which represents a 120 basis points increase relative the average from September 2014 to the day of the Contract between Lega and Movimento 5 Stelle. If we cumulate the changes in the sovereign CDS spread around political and policy dates over this period, we obtain a value of around 35 basis points. This cumulated change in our instrument can ‘account’ for a large fraction of the observed BTP-Bund spread change (approximately 70 basis points out of the 120 basis points change between the two periods), using an impulse response of around two. For a more formal assessment based on the forecast error variance decomposition, see Balduzzi et al. (2020). 

 8 The behaviour of the ISDA basis around 2018 has also been highlighted by Minnenna (2017), Ignazio Visco, Governor of the Bank of Italy in Visco (2018), Gros (2018) and Balduzzi et al. (2018b). See also Kremens (2019) and Cherubini (2019) for an analysis of the redenomination spread and De Santis (2019) for an analysis on the quanto spread.

a

A

Topics:  Europe's nations and regions Politics and economics

Tags:  Italy, populism, politics, political risk, Credit Default Swaps

Professor of Finance, Boston College

Assistant Professor of Economics, Sapienza University of Rome

PhD student, Boston College

Professor of Economics, Boston College; Research Fellow, IZA

Events

CEPR Policy Research